INNOVATION OF CREDIT RISK MEASUREMENT OF LISTED COAL COMPANIES BASED ON FSVM-KMV

China has implemented supply-side reform policies such as reduction of production since 2016, credit risk of coal industry enterprise increases. Under such circumstances, a KMV model is used to measure the credit risk of listed enterprises in coal industry and the fuzzy support vector machine (FSVM) is introduced to excavate and depict the change of enterprise value, which overcomes the limitations of small sample, non-normal distribution and subjective credit rating. We analyze the influencing factors of enterprise value and its sensitivity to the distance default of these enterprises in order to ensure the mechanisms for change in default of coal enterprises currently. To guarantee the accuracy of the empirical analysis, 10 enterprises with credit grade above AA+ with raising trend and 10 below AA+ with decline trend were selected as the experimental group and the contrast group respectively. The empirical results show that, FSVM-KMV integration model can effectively handle the nonlinear relationship between variables. It improves the prediction accuracy of enterprise credit change. This turns out that proportion of main business revenue (PBR), total return on assets (ROA) and asset-liability ratio (AD) are the main influencing factors of enterprise value in coal industry. Thus, enterprise credit is changed accordingly. Meanwhile, the empirical results of the model are basically consistent with the Standard & Poor's rating results. Therefore, the integrated model shows strong adaptability.